Detection of sulfonamide resistant bacteria and resistance genes in soils

Manure application could accelerate the environmental dissemination of antibiotic resistant bacteria (ARB) and antibiotic resistance genes (ARGs) in soils. In this study, the prevalence of sulfonamide resistant bacteria and resistance genes was investigated in agricultural soils to which organic manures had been applied in Tianjin, China. Anti-sulfonamide bacteria were found in the range of 3.29 × 104 to 1.70 × 105 CFU/g dry soil, occupying 1.5% to 2.2% of total viable counts. And sulI and sulII genes were detected in all sampling sites, with relative abundances of 5.69 × 10−5 to 6.95 × 10−4 and 4.28 × 10−4 to 1.25 × 10−3 respectively. No significant correlations between cultivable sulfonamide resistant bacteria and sul genes were found in this study. While sulI showed significant positive correlation with soil organic matter. Overall, the results highlight that soil plays an important role in resistance genes capture as the environmental reservoir.


Introduction
Manure is a major source of antibiotic pollution in the environment. Since many antibiotics are poorly absorbed by the animals and subsequently are excreted into the environment by urine or feces. The antibiotics tend to persist and accumulate in soils after repeated manure application. Residual antibiotics may exert selection pressure on environmental microorganisms, accelerating the environmental dissemination of antibiotic resistance. Bacteria have been shown to readily share genetic information by horizontal gene transfer (HGT) mechanism driven by mobile genetic elements, permitting the transfer of antibiotic resistance genes (ARGs) among environmental bacteria. Numerous antibiotic resistant bacteria (ARB) and ARGs have been widely detected in soil in the past few decades [1][2][3][4]. ARB and ARGs in soils have the potential to pose risks to human health, as susceptible pathogenic bacteria can become resistant by acquiring resistance genes in the environmental media [5]. Therefore, it is of great importance to investigate the behavior of ARB and ARGs in agricultural soils.
Sulfonamides are one of the most commonly used in livestock system. Sulfonamide-resistance genes (sul genes) were also observed with high levels of abundances in Tianjin area [6]. In this work, the prevalence of sulfonamide resistance was investigated in agricultural soils to which organic manures had been applied from four districts in Tianjin, China. Culture-dependent method was used to assess the resistance rate of bacteria in soils exposed to different sulfonamide concentrations. The quantitative PCR (qPCR) was used to quantify sulI and sulII genes. Meanwhile, the relationship between ARGs, ARB and environmental factors was also investigated.

Soil samples
Soil samples were collected from agricultural fields to which organic manures had been applied. The samples were taken from Jixian, Xiqing, Wuqing and Dagang Districts (define as JX, XQ, WQ and DG respectively) in Tianjin, China, between March and May 2016. For each site, three subsamples were taken from 0 to 10 cm surface soil and were mixed to form one composite sample. Fresh samples were processed immediately for cultivation of ARB. The other samples were stored at -80°C before DNA extraction and chemical analysis. Weights of wet soils and oven-dry soils were measured to derive soil moistures. Soil pH was determined with a soil to water ratio of 1:2.5. Total carbon (TC) was measured by a by a TOC analyzer (TOC-VCPH, SHIMADZU). Organic matter (OM) was determined by the K 2 Cr 2 O 7 oxidation method. The physicochemical properties of soil samples are summarized in Table 1.

Detection of ARB
Each soil sample was measured in duplicate to determine the numbers of ARB using colony forming units count method described by Chen et al. [5] with minor modifications. Briefly, 5.0 g of wet soil was mixed with 45 mL sterile saline solution (0.85% NaCl) and shaken vigorously on an oscillator at 200 rpm for 20 min. 100 μL of ten-fold serial dilutions for each suspension were spread onto a broth agar plates containing selected concentrations of antibiotics (20 and 50 μg/mL of sulfadiazine). Broth agar plates with no antibiotics was used to determine the total cultivable bacteria numbers.

Detection of ARGs
Total DNA was extracted from 0.5 g of soil (fresh weight) by using FastDNA SPIN kit for Soil (MP Blomedicals, LLC., France). The concentration and quality of the extracted DNA was determined by spectrophotometer analysis (NanoDrop ND-1000, Theromo Fisher Scientific). PCR assays were used for broad-scale screening of the presence/absence of sul genes, and to get the standards for subsequent qPCR analysis. All PCR assays were conducted in a Peltier Thermal Cycler (Bio-Rad). Primers and annealing temperatures are described in Table 2.
After PCR amplification, gel slices of an agarose gel containing the PCR products were excised, and purified using EasyPure Quick Gel Extraction Kit (TransGen). The purified PCR products were ligated into p-GEM T easy vector (Promega) and then cloned into Escherichia coli DH5α (Tiangen). Clones containing target gene inserts were picked and sequenced. If the gene inserts were verified as the object resistance genes using the BLAST alignment tool, clones that had right gene inserts were chosen as the standards for the subsequent qPCR. Plasmids carrying target genes were extracted with Plasmid Kit (TaKaRa). Two target genes (sulI and sulII) were quantified by qPCR using a SYBR-Green approach. 16S rRNA genes were quantified according to the TaqMan qPCR method [8]. 10-fold serial dilutions of a known copy number of plasmid carrying respective genes were generated to produce the standard curve. The qPCR mixtures (total volume, 20 μL) consisted of 10 μL of SYBR    figure 1, total concentrations of cultivable bacterial cells in soils ranged from 2.05 × 10 6 to 7.86 × 10 6 CFU/g dry soil. Cultivable live anti-sulfonamide bacteria were observed in all soil samples, ranging from 3.29 × 10 4 to 1.70 × 10 5 CFU/g dry soil. The relative abundances of sulfonamide-resistant strains (resistant to 50 μg/mL of sulfadiazine) observed in this study (1.5% to 2.2%) were much lower than those in wastewater-irrigated soils (approximately 10% to 20%) [ These resistant bacteria may come directly from the indigenous soil environment. They may also result from the stress of antibiotics in soil which needs to be further studied. Meanwhile, the sulfonamide resistance rate of bacteria in soils exposed to different sulfonamide concentrations are investigated ( figure 2). With the increase of sulfonamide concentration (from 20 to 50 μg/mL of sulfadiazine), the resistance rate of bacteria showed a decreasing trend (4.4%-9.4% at 20 μg/mL, 1.5%-2.2% at 50 μg/mL). Significant difference was observed in resistance rate under different sulfonamide concentrations (P < 0.01). But exposed to the same sulfonamide concentration, there is no significant difference in drug resistance of bacteria among different samples. The sulfonamide resistance rate of bacteria in soil samples exposed to different sulfonamide concentrations (20 and 50 μg/mL of sulfadiazine).

Determination of sulfonamide resistance genes
Absolute gene copy numbers of two sul genes (sulI and sulII) in soil samples were presented in figure  3a. The absolute abundances of sulI genes in JX (1.58 × 10 7 copies/g dry soil) and XQ (1.82 × 10 7 copies/g dry soil) were comparable, but significantly higher than those in WQ and DG (10 6 copies/g dry soil). The highest abundance of sulII genes was occurred in WQ (1.32 × 10 8 copies/g dry soil), approximately 1 order of magnitude higher than those in other three samples. This could be related to the high 16S rRNA gene copies in WQ (1.58 × 10 11 copies/g dry soil). To minimize variance caused by different extraction and analytical efficiencies and differences in background bacterial abundances, the absolute number of all resistance genes were normalized to that of ambient 16S rRNA genes (figure 3b). The same general trends in gene abundance were seen in the normalized data relative to absolute data, since the total number of 16S rRNA gene copies was found to be relatively consistent among different sites at 10 10 copies/g dry soil, except for WQ. The relative abundances (target gene/16S rRNA genes) of sul genes showed significant variation over sampling sites, ranging from 10 -5 to 10 -3 . In all sampling sites, the relative abundances of sulII (4.28 × 10 -4 to 1.25 × 10 -3 ) were higher than those of sulI (5.69 × 10 -5 to 6.95 × 10 -4 ). This observation is consistent with the results obtained in the aquaculture environment of Tianjin (relative abundances of 10 -4 to 10 -3 for sulII, and 10 -5 to 10 -4 for sulI) [6]. The total absolute and relative abundances of two sul genes (sulI and sulII) on average in different sites were in the range of 3.70 × 10 7 to 1.41 × 10 8 copies/g dry soil, and 8.91 ×10 -4 to 1.95 ×10 -3 respectively.
In this study, sulI and sulII genes were detected with high abundance, which were comparable to those in wastewater and reclaimed water irrigated soils in China [4,5]. The high prevalence of sul genes could be related to residual antibiotics in soils introduced by manure application or wastewater irrigation. However, even in the absence of continuous selection of sulfonamides, sulI was found to be associated with persistent sulfonamide resistance in microorganisms [6]. Overall, the results highlight that soil plays an important role in resistance genes capture as the environmental reservoir.   Table 3. Correlation analysis of relative abundances of ARGs, ARB and environmental factors ** represents statistical significance with P < 0.01; sul represents the sum of sulI and sulII genes.

Relationship between ARGs, ARB and environmental factors
According to the Spearman's correlation coefficients (Table 3), no significant correlations between cultivable sulfonamide resistant bacteria and sul genes were found in this study. This is not surprising given that only a very small fraction of soil microbes were targeted in cultivation. A considerable part of sul genes quantified could possibly exist in uncultivable bacteria. In addition, the relative abundance of sulI showed significant positive correlation with soil organic matter. This observation was in agreement with previous report [3]. But for sulII, no significant positive correlation was found with organic matter. Recent data have shown a significant negative relationship between ARGs and soil pH, since soil pH exerted a strong selection pressure on soil microbes and appeared to have a